‘The right man in the right place’ – the consequences of gender-coding of place and occupation in collaboration processes

Society needs to find new ways to utilise its resources in the best possible way in order to enable satisfactory services for its citizens in the long term. This is particularly important in sparsely populated areas, and in cities and municipalities with a declining population. This study contributes to this field by analysing a project for collaboration between the rescue service and the home-care service that has been introduced in a number of Swedish municipalities. The collaboration is intended to ensure welfare and safety for citizens, to guarantee a more efficient use of municipal resources, and to contribute to improved emergency management and civil protection. The rescue service and the home-care service are two clearly gender-coded occupations that also operate on gender-coded workplaces and places of work. An overarching aim has therefore been to study gendered obstacles and possibilities of the collaboration. In our analysis of the empirical data – interviews with persons involved in the collaboration – place emerged as an important aspect of the collaboration processes, and is therefore elaborated in this article to contribute with knowledge of how conceptions and gender-coding of places and occupations affect sustainable and well-functioning collaboration processes. The results show that collaboration processes between municipal services are complex and challenge ideals of the organisation, content and responsibility of work and who should perform certain work tasks. Notions of gender and gender differences are reproduced through the collaboration, which affect the efforts of municipalities to ensure welfare and safety for citizens.


Introduction
Society needs to find new ways to utilise its resources in the best possible way in order to enable satisfactory services for its citizens in the long term. This is particularly important in sparsely populated areas, and in cities and municipalities with a declining population. In Europe, one-third of cities are affected by population decline, leading to financially demanding situations (Carbanaro et al., 2018). Even though these are well-known challenges, there is a lack of research offering substantive guidance for a regional policy as well as political strategies to handle shrinkage (Carbanaro et al., 2018;Syssner, 2016). This study contributes to this field by analysing a project for collaboration that has been introduced in a number of Swedish municipalities to ensure welfare and safety for citizens, to guarantee more efficient use of municipal resources, and to contribute to improved emergency management and civil protection. The rescue service and the home-care service are the collaborating parts in this project, where the home-care service is expected to perform certain tasks in the event of an emergency while waiting for the rescue service, and the rescue service is expected to work on prevention together with the home-care service. The studied collaboration project could be an important contribution to possible strategies to tackle population decline, in terms of both financial and personal resources.
Collaboration, by which we here mean interaction over organisational bordersin or between municipalitieswith the aim to solve a specific issue, has increased as a measure to secure competence provision and maintained service (Berlin and Carlström, 2011;Lindberg, 2009). Also in the cities, more collaboration between different sectors has been initiated to increase social and economic sustainability as resources are used more efficiently, as to for example conduct emergency operations more quickly.
However, even if collaboration is often delivered as a 'universal solution' for efficient use of resources, research shows that these processes are not always as easy as they sometimes seem in a project plan. Huxham (2003) describes that collaborations in practice often end up in inertia, even though an advantage that could not be reached by one organisation alone has been the purpose of the collaboration. A well-functioning collaboration can have great advantages, but the pain required to get there needs to be worth it. The processes are often difficult and complicated to implement in an efficient and satisfactory way, as organisations need to be willing to change their ingrained and traditional working methods and forms of organisation in order to try new methods, forms of collaboration and partners (e.g. Berlin and Carlström, 2011;Huxham, 2003;Löfström, 2010;Stenberg, 2009). Just like other implementation processes of organisational change, successful collaboration processes require for instance good communication, clear distribution of tasks and responsibilities, a clear idea of the common goal and balanced power relations between the parties involved (Meyers et al., 2012;Stenberg, 2009;Weinholt, 2010). Resistance can be a consequence of insufficient implementation, that is if components important for the realisation are overlooked. It is therefore important to give all actors involved in the process of change capacities to contribute, but also to request results and hold people responsible for their actions (Benschop and Verloo, 2006;Lombardo and Mergaert, 2013). The collaboration processes we analyse in this project do not explicitly aim to change gender-coded structures and cultures, but change comes as a consequence of the implementation of the collaboration.
This study was initiated by the Swedish Civil Contingencies Agency (MSB) with the aim to look into gendered obstacles and possibilities of collaboration between the rescue service and home-care service. In our analysis of the empirical datainterviews with persons involved in the collaborationplace emerged as an important aspect of the collaboration processes, and is therefore elaborated in this article to contribute with knowledge of how conceptions and gender-coding of places and occupations affect sustainable and well-functioning collaboration processes. Furthermore, too little attention has been paid to the ways place affect the doing of gender in organisational studies (e.g. Tyler and Cohen, 2010;Wasserman and Frenkel, 2015). Our study therefore contributes to gender studies of workplaces and places of work, which means that we will discuss places where labour efforts are carried out, in addition to the workplace itself.
In the following section, our case is introduced, placing the collaboration and the rescue service and home-care service in the gender-segregated Swedish job market. Research on collaboration involving rescue services and home-care services is also discussed, showing that our study contributes a link between research on collaboration processes and research on the everyday practices of workers.
Our theoretical framework is thereafter introduced, discussing the relations between gender, place, and work, followed by a methods section. After this, our empirical data is presented and analysed from four identified important places for the collaboration, where gender and the gender-coding of these places constitute an obvious obstacle for well-functioning collaboration. In the conclusion, the obstacles are set in relation to structural conditions. Finally, we return to the question of gendered obstacles and possibilities for collaboration processes and the knowledge this study brings to practical work to ensure economic sustainability as well as welfare and safety for citizens.

Introducing the case
In Sweden, both the rescue service and home-care service are publicly financed services that operate in the local arena. They provide services that need to be supplied to all citizens wherever they live. In sparsely populated areas or municipalities with population decline, this can prove a challenge, and people might have to wait (too) long for the fire fighters or caregivers to arrive in case of emergency. More than 40 percent of Swedish municipalities have seen a demographic decline in the last decades, at the same time as the public sector cost per inhabitant has increased (Karlsson and Lundholm, 2014), which points to the urgency of finding creative solutions. Collaboration between different sectors in municipalities has increasingly been seen as a solution and important tool to ensure welfare and safety for citizens, as well as to promote economic sustainability in municipalities (Lindberg, 2009). In this project, the collaboration between rescue services and home-care services is in focus.
The rescue service and the home-care service are two clearly gender-coded occupations, and typical examples of gender segregation that exists in the entire Swedish labour market, where occupations and occupational fields dominated by men or women create unequal conditions in and between occupations (e.g. Holt et al., 2012). The rescue service is male-dominated and characterised by notions of masculinity in different ways (e.g. Ericson, 2014;Khan et al., 2017), compared to the home-care service which is female-dominated and characterised by notions of femininity (e.g. McDowell, 2009). In Sweden, seven percent of the staff within home-care service, home nursing and nursing homes are men (SCB, 2016) and barely four percent of the employed firefighters are women (MSB, 2016).
Research on collaboration processes where the rescue service is involved shows that the collaboration rarely stretches beyond the actual rescue operation and only to a limited extent contributes to changes in the standard routines of the rescue service. Berlin and Carlström (2011) describe how the collective image of always being in the first ranks can hinder an effective collaboration and instead build hierarchies and territorialities. Weinholt (2010) shows that it is mainly the other party involved in the cooperation that is affected through new/additional tasks. Several cases of reluctance within the rescue service to integrate new work models and tasks, with references to the core mission of the rescue service, can be found (e.g. Yosefi Mojir and Pilemalm, 2014). A research project on nurses' experience of a collaboration project with firefighters shows that the nurses experienced some difficulties and issues of concern regarding the project, for example, role confusion, problems combining emergency care with their regular work tasks, setting priorities, as well as a lack of group belonging and feedback (Svensson and Weinholt, 2014). When it comes to research on collaboration and the rescue service and/or the homecare service, the absence of a gender perspective is striking, both as a research question and in the analysis. However, a research overview of gender perspectives on management and collaboration in the event of an accident or crisis (Jansson and Linghag, 2015) emphasises that the field is characterised by men and masculinity in an unproblematised way. This has consequences for risk and needs assessments, the adjustment of bureaucratic organisations in emergency events and the handling of the temporary forms of organisation that are created in these situations. The overview also shows that notions of gender affect collaboration between different parties, particularly between the rescue service and the partners they work with. Our research takes these conclusions one step further, focusing on the collaboration processes per se, rather than the specific event of an accident or crisis. The article, therefore, contributes knowledge on the everyday practices and places of collaboration processes from a gender perspective, and thus fills an important gap in the field surrounding collaboration processes.

Gender, place and work
The fact that work is gender-coded is well-established. Different occupations, positions and work tasks are gendered and marked, that is, assumed to be more or less suited for a certain gender (Acker, 1990), and the labour market is likewise a part of creating and recreating gendered identities (McDowell, 1999(McDowell, , 2004. In other words, the labour market is formed by perceptions of gender-specific 'suitable' tasks, but it is also a part of forming identities. Also within organisations notions on whether or not women or men are suited for a certain occupation or certain work tasks create segregation. By studying the structure and culture of an organisation, conceptions of and presumptions about men's and women's positions and relations as consequences of the gender order of that organisation are made visible (Wahl et al., 2018). Focus is in this way transferred from individual characteristics and abilities to the organisational gender order that often conditions the positions of women and men in a workplace and within an occupation. It is a persistent pattern that is repeatedly reproduced, even when organisations go through organisational change (Acker, 1990). To these well-established ideas, this article brings understandings of organisational place. Research on the material position in a workplace (e.g. Dale and Burrell, 2008;Hofbauer, 2000) shows that the structure of an organisation can often be understood by the positioning of workers in the workplace, the internal built environment of the organisation. However, these studies often lack a thorough gender perspective. Therefore we argue that more research is needed that combines the three perspectives of gender, work and place (see also McDowell, 2004). From an active interest in gender, work and workplaces in the 1980s and 90s, with titles like Women, Work and Place (Kobayoashi et al., 1994) and Gender, work, and space (Hanson and Pratt, 1995), the interest in these issues seems to have faded (see also McMorran, 2012). Some important exceptions of course exist, with Linda McDowell (e.g. 2004McDowell (e.g. , 2009McDowell (e.g. , 2012 at the front. However, we argue that the gender-coding of work and the gender-coding of place need to be understood as a unity to truly understand the gender division of labour (Johansson and Lundgren, 2015;Kobayoashi et al., 1994;Tyler and Cohen, 2010), and therefore line up with those trying to return the human geographical interest in the gendered workplace.
A few studies from the two last decades have used theoretical perspectives and concepts from human geography to approach the gender segregation of workplaces. Wasserman and Frenkel (2015) use Lefebvre's theory of the production of space, and argue that even though architects have an important role in (re)producing organisational inequalities by mere design, the daily experiences of employees and the lived spaces that they inhabit are important to understand organisational order. They show that both gender and class determine how persons experience and operate within a certain organisational space. Wasserman and Frenkel's study therefore points to 'the importance of understanding the organizational space as a part of any organization's inequality regime' (2015:1500, Dale and Burrell, 2008). This is also the result of Halford and Leonard's (2003) study that concludes that both the material and the symbolic construction of hospital places and spaces shape the experiences, performance and identity construction of nurses. Johansson and Lundgren (2015) come to similar conclusions in their study of a supermarket. They show how spatial divisions and boundaries were incorporated into the practice of gender and work. McDowell (2004) also points out that both the location and the layout of a workplace reflect power relations, and shape different workplace behaviours. The workplace is not an empty container waiting to be filled with gendered relations but has an active influence on gender-codings (see also Tyler and Cohen, 2010). Simpson's (2014) study of male cabin crew workers is a good example of this. The masculine-coded cockpit, symbolised by technology, uniforms, rationality, danger and expertise, stands in contrast to the feminine-coded cabin, symbolised by service, consumption, trivia and entertainment. Simpson shows how places are gendered but are also gendering, and how the male cabin crew became marked by the feminine-coded place they operated in. From this, we argue that the gender-coding of workplaces is an important factor to bring into collaboration theory and understandings of implementation of organisational change. Geographical locations, places for cooperation and physical co-localisation, are factors of great importance for different types of collaboration processes (see also Löfström, 2010). Through these places, boundaries can be created, contested, negotiated or erased (see also Tyler and Cohen, 2010).
In our study, the workplace is not one place, but many. Both the rescue service and the home-care service have a workplace, a sort of basecamp, from within which they depart and to which they return after performing many of their work tasks in other places. We will therefore also talk of the places of work. The places of work are gendercoded places too, bearing the stamp of the classical divide of private and public. Including different places of work, this study also contributes to calls from health geographies to engage with health workers and their places of work; for instance communities, workplaces and homes (Andrews and Evans, 2008;Connell and Walton-Roberts, 2015).

Method
This study took the form of case studies in a selection of municipalities where some kind of collaboration processes between the rescue service and the home-care service had been set up for testing. Eight municipalities of varying sizes were included in the study. They had somewhat different ways of defining the collaboration, and there were also differences in how far they had come in the process. Yet, the processes studied share many common characteristics. The tasks which the nursing staff perform, or are expected to perform, within the collaboration process include for example responding as the first resource in a 'call-out'. In some of the municipalities studied, the nursing staff are expected to carry out fire inspections in clients' homes. A common task for the rescue service in the studied collaboration processes is to respond to care alarms and calls which involve lifts if a client has fallen and the nursing staff for different reasons are unable to lift the person.
The study is based on interviews with 12 women and 18 men. We have categorised the interviewees as either 'rescue service staff' or 'nursing staff'. The group 'rescue service staff' includes managers (at different levels), firefighters and fire engineers, while the group 'nursing staff' includes home-care nurses and managers (at different levels). In total, the group 'rescue service staff' consists of 20 people and the group 'nursing staff' consists of 10 people. The rescue service staff in the study describe their work as primarily being about 'responding to call-outs' or 'conducting operations'. This refers to emergency calls in the event of a traffic accident, fire, cardiac arrest (pre-ambulance) and so on. In these call-outs, the response often includes different levels of medical care. The rescue service staff also mention the prevention work they do to prevent different types of accidents in people's homes and public places. The nursing staff in this study describe their work as being primarily about planned visits to clients' homes to take care of routine tasks such as administering medication, helping clients to get out of bed, use the bathroom and so on. But the work tasks also include unforeseen and sometimes acute events like responding to a safety/care alarm (e.g. that a client has fallen and needs help to get up, or more critical situations like a client having suffered a stroke or a heart attack).
The gender-coding of place was not a part of the initial research questions of the project behind this study, and was therefore not included in our interview guide. In the analysis of the interviews, place however turned up as important as obstacle and possibility for cooperation. For this article, we have focused on those aspects of the interviews. Our methods therefore have similarities with analytical processes in Grounded Theory, without being a 'proper' or regular GT analysis. In Grounded Theory, the initial aim is broad and open, and becomes more and more focused along the process of analysis, and questions raised during the analysis are allowed to guide the aim and the data chosen. The contextualisation in relation to research is a part of the final stage of the analysis (Bryant and Charmaz, 2019).
Our analysis of the material was done through repeated readings discerning recurrent themes and patterns, and usage of memos (Layder, 1998). This resulted in a number of central and recurring themes, such as for instance status, understandings and experiences of the collaboration, everyday work, and resistance. We also looked for gender-codings, and how obstacles and possibilities were expressed in the interviews. These themes formed the basis for our analysis and report to the Swedish Civil Contingencies Agency (MSB) who wanted answers to questions about the obstacles and possibilities of the collaboration. In our discussion, we came to the conclusion that place seemed to matter, which has received special attention in this next step of the analysis. We have returned to the interviews once more to look for how and what places find expression, and analysed that result with a gender perspective.
The presentation and discussion of the empirical material are structured by the different places of work that have come up in the interviews. These include the fire station, the car, the scene of the accident and the home. All these four places are clearly gender-coded, and also important places for the collaboration process. The fire station and the car are traditionally masculine-coded, while the home is traditionally feminine-coded. The gender-coding of these different places becomes a central factor in enforcing and upholding boundaries between rescue service staff and nursing staff. The scene of an accident could be labelled a more neutral place. However, in our material, it is put in contrast to the home. The boundary between private and public places proves to be important in upholding the status of the rescue service.

The places of the cooperation
The fire station The fire station is an obvious example of a gender-coded place that is filled with meaning even before the collaboration begins. It defines who self-evidently belongs to the place and who does not, and how one should act in this place (McDowell, 1999). As will be shown, the fire station is primarily seen as belonging to the rescue service staff, even in the cases with co-localisation, where the nursing staff are stationed or have access to the fire station, and where active efforts have been made to make it a commonplace for both groups. For both groups it is evident that the fire station takes an important symbolic meaning. One example is that the nursing staff experience a higher status when they are based at the fire station.
We are not pushed aside anymore […] When we say that we are based at the fire station, they're like oh, are you up there? It sort of raises the status. (Nursing staff, home-care service) The fire station is a masculine-coded place and its status and reputation reflect onto the nursing staff when they are allowed to be associated with this place (cf. Simpson, 2014). In addition, the nursing staff in many ways benefit from the co-localisation, for instance, because of access to a free gym at the workplace, and well-functioning buildings as their 'base'.
A few of us [nursing staff] use the gym.
[…] Some months ago, I was there and ran on the treadmill, the one in the middle. Two firefighters came too, one on each side of me. Then I felt that, no, it's time to stop. But it is not about them, only your own body issues. (Nursing staff, homecare service) Co-localisation at the fire station does not mean that the two groups automatically interact, as has been intended. Practically, we can see that the boundaries are often actively upheld at the fire station to separate the two categories of staff. This is done both spatially and through interaction. Spatially, nursing and rescue service staff often have different 'bases' at the fire station. These 'bases' are only used by one of the staff categories. There are also common areas. In these places, the boundaries are drawn through interactions that in different ways uphold the boundaries. This is manifested through examples like the rescue service staff closing their door when the nursing staff arrive at the fire station on their breaks, or intra-collegial complaining about the other group.
Sometimes there is some muttering [over the nurses], like 'Didn't they bring out the waste', 'Aren't they gonna do the dishes'. (Rescue service staff) Occasionally there are confrontations where the nursing staff are told off by the rescue service staff for something they have or have not done, in the common area. These become actions which uphold power hierarchies, and which mark who has the power and sets the rules in the commonplace. Even among those who do not describe active dissociation, it is stated that one rarely exchanges more than a brief 'hello' in the corridor. The fire station is an interesting case here when talking about gender-coded workplaces. It is a workplace, but it is also like a second home for firefighters. As part of their work schedule, they sleep there, make dinner, work out, and watch TV. Moreover, the firefighters take care of the fire station like a private home; cleaning and doing the dishes. We interpret this combination of private and public in one and the same place as one of the reasons why the nursing staff are kept at a distance. It is important for the rescue service staff to keep a notion of a homosocial male place, to prevent the feminine-coded nursing staff from 'infecting' the fire station with too much of private home codings (cf. Simpson, 2014).

The car
Cars also appear as both an obstacle and a possibility for collaboration. In several of the municipalities studied, the nursing and rescue service staff share one or more cars. The nursing staff say that they get access to much better cars when they share them with the rescue service staff, and thus they view this as a positive aspect of the collaboration.
We need better cars. Ours are not equipped with what we need. There are no extra lights, we drive at night… They are not equipped for our work. We are not allowed to choose our cars, they are assigned us. The ones we share with the rescue service are better cars really. I don't understand the logic. The rescue service can buy new cars, but we can't. (Nursing staff, homecare service) The cars of the home-care service are, as in the quote above, described as impractical and unsuitable for the needs of the nursing staff; they do not have four-wheel drive, there are no extra lights, they are low and small. The cars shared with the rescue service staff are, by contrast, reliable cars with four-wheel drive, and therefore contribute to better mobility in the nursing staff's occupational practice.
However, shared cars also create a basis for conflict between the two occupational groups, which contributes to a reinforcement of the boundaries between them. Many of the conflicts refer to the maintenance of the cars, both the shared ones and the ones belonging to the home-care service solely. The rescue service staff argue that the nursing staff do not take care of the cars correctly: They're dirty as hell and I don't know how they drive, but sure, we can wash them. As long as it's just the outside, I won't touch the inside. (Rescue service staff) Cars are traditionally connected to masculinity. However, just as place is both gendered and gendering, so are cars. Cars shape their users but are also shaped by them (Balkmar, 2012), which can explain why the rescue service staff do not want to handle the cars of the home-care service. They are not the kind of cars that the firefighters want to be associated with. The nurses also mention that they do not always take good care of the cars, but explain this behaviour with a logic related to the work practice. Some nurses drive without taking care, you might hit a curbside. You want to come as close as possible to the care-takers, so it can happen that you hit a lamppost or something. It can explain why the fire fighters don't want to share cars with us. (Nursing staff, home-care service) Previous research describe clear routines, rules and roles regarding tasks as an important factor for successful collaboration processes (e.g. Berlin and Carlström, 2011). We have found that this is often lacking when it comes to the common cars, or at least that it is an area where the rules are not well established. How and why the maintenance of the cars works the way it does is not always clear to the co-workers. There seems to be no clear division of the work tasks between the two groups involved in the collaboration, which shows in the discussion about car maintenance. The home-care staff can see that rescue service staff take care of their own cars during work time, so why not maintain the cars of the home-care crew too, since they do not have time allocated for this task. The rescue service staff on the other hand view car maintenance as a part of workmanship and preparedness for call-outs, which should not be mixed up with general vehicular service. An ongoing discussion about the maintenance of the cars contributes to division and conflicts between the groups.

The home
The homethe private place in the classical divide between public and private placesis another example of a gender-coded place that is filled with meaning already before the collaboration begins. In this study, the home as a place for nursing activities is reinforced in the interviews. The rescue service staff clearly express that they do not view homebased nursing activities as part of their job unless part of an acute accident. I'm not here to take care of old people, I mean it's the women who are employed to do that. Then let them do it. (Rescue service staff) They [the nurses] do the nursing stuff in homes, the caregiving part. We don't have that caregiving part in our work. (Rescue service staff) The gender-coding of the occupations becomes even more evident, however, when the collaboration is discussed with the rescue service staff based on questions of status. While the collaboration processes might lead to a higher status for the nursing staff, the rescue service staff fear that the reverse would apply to them; that their occupation would assume a lower status by taking on too much of the nursing role and/or being associated with work tasks which to a great extent are performed by women and in private homes. They do not want tasks that include too much nursing, particularly responding to care alarms, which means caring for someone in their home. This is something which the rescue service staff have been doing already before the collaboration processes started, but there is a concern that the collaboration will make it a more frequent element in their work routines.
We respond to quite a few care alarms these days, but if there will be too many caring activities [in the rescue service], I believe that people will lash out, like we are not employed to.., this is sort of not my occupation. (Rescue service staff) As we can see, the home is rarely mentioned as place, but the work carried out there is instead described as a practice of care that takes place in private homes. The division and gender-coding of public and private places here become a barrier and marker for where different caring activities can be performed and by whom (McDowell, 1999). Just as the gender-coded home-care service could 'infect' the fire station, we interpret the firefighters' resistance towards caring activities in private homes as a fear of being gendered by place, like the male cabin crew Simpson (2014) describes.

The scene of an accident
Nursing activities that are part of an emergency operation in public places, for example at a car accident, fall within the framework of what the rescue staff describe as their tasks. In several cases, like in the quote below, these are also described as places where nursing staff do not belong. Now we've found out that they [the nursing staff] are going to be called out for stuff like FIR [first rescue resource] and in our opinion that's completely screwed up. […] [T]hey're not trained to do it. The training they get is not good. There are ladies who are over 50 years old, are they going to arrive and close off a road, perhaps throw in extinguishing grenades? On the motorway? When we're on the motorway, we have large trucks to close off in order to protect ourselves and here they are with a small bloody car. If anything, it could lead to more accidents. They have their tasks, they should let them do that instead […] If they were to arrive a minute before us, it could of course be the case that they might save a life by supporting someone's neck but they might injure themselves too. That's how we feel. It doesn't feel right. (Rescue service staff) Here, the nursing staff are described as unhelpful (besides perhaps saving someone's life), making things more difficult for the rescue service staff and also exposing themselves to danger. The gender-coding of the nursing staff, their cars, and the place of accident are important in creating distance and in describing the collaboration as unhelpful; the (middle-aged) women do not belong to this place and it would be better if they stayed with their work tasks inside the homes. On the other hand, the nursing staff describe themselves as a natural part of public places. In transporting themselves between different caretakers' homes, they take place in large areas of the municipality.
The fire fighters are not out on the streets at night, but we are. We are maybe the ones that are most often out on the streets, in the whole municipality. They [the employer] want to make use of what we can do. In the end, we are trained in health care, more than the fire fighters are. (Nursing staff, home-care service) It can also be the case that we see a fire… It has actually happened that we saw a place on fire, and called the rescue service. We will have extinguishing grenades, so maybe in the future we will be able to do something too. (Nursing staff, home-care service) It should be noted that there are also rescue service staff that see advantages with the nursing staff being the first to arrive at the scene of an accidentthat they can 'do many little things', 'warn', 'help us [with medical care]', and guide the firefighters to the right place based on better local knowledge: The only thing I can think of, where they can help us out, is to tell us the exact address, because in the countryside, it can be hard to find the way. (Rescue service staff)

Conclusion: The consequences of gender-coding for effective collaboration
This study shows that the collaboration processes between municipal services are complex and challenge ideals of the organisation, content and responsibility of work and who should perform certain work tasks. We have been able to illustrate how notions of gender in male-dominated and female-dominated organisations are both reproduced and reinforced through assumptions about specific work tasks for each of the organisations and those who work there, but also the places they operate in. The fire station is an obvious example of the 'inequality regime' described by Wasserman and Frenkel (2015) where gender-coding of occupation divides the place and where the nursing staff are not allowed equal right to belong as the rescue service staff (see also McDowell, 1999). Since both occupations operate in other places than their 'home base', we can also show that the 'inequality regime' extends beyond the workplace, based on the classic divide of private and public places. The fact that the rescue service staff provide care at the scene of an accident, but preferably not in private homes, shows that the gender-coding of private and public place frames the understanding of where different activities should take place. Understanding and dealing with the gender-coding of occupations and places is therefore an important condition for accomplishing collaborative tasks.
Further, the study shows that gender differences are reinforced as the rescue service staff uphold differences between their work and that of the nursing staff, which includes the workplace and the places of work. This relates to the idea that collaboration processes could induce lower status for the rescue service staff if their occupation becomes 'too' associated with nursing staff and nursing tasks taking place in homes, as opposed to the nursing staff who rather consider the collaboration a boost in status. In addition, this study shows that the nursing staff do not only gain from the collaboration in terms of status. They also get better resources, like cars, a free gym at the workplace, and wellfunctioning buildings as their 'base' where they start and end their workday. At the same time, they often have to abide with being insulted by the rescue service staff, which we interpret as a kind of protectionism, that is, the reluctance of the rescue staff to let other occupational groups 'barge into' the arena which they consider theirs. Moreover, it is the main structure and culture of the rescue service that is challenged in the cooperation with the home-care service, since the rescue service has the superior position and the preferential right of interpretation in the collaboration, concerning both status and the workplace. When the collaboration involves a co-localisation, it is the nursing staff that have moved to the already defined fire station place, a place belonging to the rescue service staff. Moreover, the tasks involved in the collaboration also reinforce the structural inequalities; the role of the rescue service is to help when the home-care service staff cannot manage their ordinary work tasks (like heavy lifting), and the role of the caring staff in the collaboration is to act as first persons on an accident or fire, and then hand over to the 'experts'. The expertise of the caring staff, for example, medical care, is not emphasised as one of the strengths of the collaboration. At the same time, it is the rescue service staff that have the most negative attitude towards the collaboration processes, while the nursing staff are the ones who mainly are/will be appointed new work tasks (c.f. Weinholt, 2010).
Change processes in the rescue service have proven to be a challenge in general, based on a persistent gender order that produces and reproduces gendered understandings of the occupation, its tasks and needed capacities (e.g. Ericson, 2014;Jansson, 2012;Jansson and Grip, 2012;Khan et al., 2017;Yosefi Mojir and Pilemalm, 2014). Taken for granted structures and privileges are made visible and challenged by the collaboration process, and therefore cause resistance. Organisational change processes often affect power relationships and conflicting interests, for instance in terms of allocation of resources, influence and responsibilities. This often raises resistance and unwillingness to implement the process of change, meaning that a lack of results in organisational change can depend on the structures, cultures and complex power relations challenged, resulting in resistance and reluctance to change (Lombardo and Mergaert, 2013;Wahl et al., 2018). Our focus on the gender-coding of place and work highlights and problematises the cooperating parties' organisational cultures and structures. This relates to the hierarchy of the two occupations. As has been shown in previous research, it is important to even out the power relations between the parties involved in a collaboration process in order to achieve sustainable collaboration (e.g. Weinholt, 2010). By the time of our study, this had not been addressed in the collaboration processes.
We have shown in this study the importance of including gender analyses in collaboration theory, and of including place in analyses of gender segregation of work. Our results show that notions of gender and gender differences are reproduced through the collaboration, and that they affect the efforts of municipalities to ensure welfare and safety for citizens. Therefore we stress the importance of analysing structural preconditions and what obstacles might occurand having a plan to handle thembefore the collaboration begins.
The study is part of the project 'Sambrukhinder och möjligheter' [Collaborationobstacles and possibilities] financed by the Swedish Civil Contingencies Agency (MSB) (reg. no. 2015-639).

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/ or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: This work was supported by the Swedish Civil Contingencies Agency (grant number reg. no 2015-639).